Implementation of micromethods to resolve problems of human breast tumor heterogeneity in analysis of cyclic 3':5'-nucleotide phosphodiesterase.

Individual human infiltrating ductal carcinomas and fibroadenomas were sectioned frozen to yield an alternating sequence of stained and lyophilized material. Stained preparations were used as references permitting microdissection of regions of tumor involvement in the corresponding dried sections. Tissues quantities of 5 to 25 micrograms dry weight were incubated under mineral oil in reaction volumes of 5 microliters and analyzed for cyclic adenosine 3':5'-monophosphate phosphodiesterase (PDE). The observed affinity constants for the 27,000 x g soluble PDE from benign tumors were 4.7 and 49.9 microM, while those for malignant tumors were 6.3 and 28.5 microM. The soluble enzyme of both tumor types eluted in three peaks on DEAE-Sephacel microcolumns. Both tumor types possessed a PDE activator eluting at 350 mM NaCl, although endogenous PDE activities were unaffected by additions of either this activator or 200 microM ethylene glycol bis(beta-aminoethyl ether)N,N,N',N'-tetraacetic acid. Individual microsections of benign tumors contained total PDE levels 2-fold higher than those of malignant tumors. Homogenates prepared from pooled microsections of the same tumors possessed only one-half of the total activity. Differential losses of enzyme in various preparation schemes as well as the use of tumor samples differing in cell density were suggested to account for some of the apparently conflicting literature values for breast tumor PDE.